Quantitatively, albumin is the most important protein in human plasma and plays a fundamental role in the maintenance of the colloidal osmotic pressure of the blood. Albumin also has other properties, such as its antioxidant and anti-freeradical activity and also its affinity for union with various endogenous and exogenous substances, such as lipids, fatty acids, bile salts, drugs and toxic substances, amongst other ligands.
The therapeutic use of albumin has been indicated, according to several authors, as a substitution therapy in patients with serious albumin deficiency, hypovolaemic syndrome or shock (serious burns, trauma, or haemorrhages), hypoproteinaemia due to chronic renal diseases and cirrhosis of the liver and also in cases of cardiopulmonary bypass, acute respiratory distress syndrome, haemodialysis and hyperbilirrubinaemia. Albumin is also used in the treatment of ascites, acute nephroses, acute nephrotic syndrome, pancreatitis, intra-abdominal infections and acute liver insufficiency.
As the principal transporting molecule in the blood, serum albumin has specific binding sites for lipophilic substances such as fatty acids, bilirubin, etc. The majority of albumin ligands bind to one of the two binding sites I or II. Free fatty acids, metallic ions such as copper and bilirubin bind selectively to specific binding domains.
Normally these substances are transported to the liver by the albumin, where they are degraded. In cases of liver failure these toxic substances stay bound to the albumin saturating its binding capacity, and producing an increase in the levels of toxins in the blood, which accumulate in the tissues and cause general multiorgan failure which can cause the patient's death. In these cases treatment can be carried out by extracorporeal liver support systems which benefit from the binding capacity of commercial therapeutic albumin. Having an albumin with greater binding capacity could be more effective in this type of treatment.
The capacity of albumin to capture and transport a large variety of substances, such as metabolites, metals, toxins, fatty acids, hormones, etc., makes this protein very important.
The processes for obtaining albumin, from human plasma normally start with Fraction V (FrV) of the alcoholic fractionation of plasma according to the Cohn method (Cohn et al., J Am Chem Soc, 1946, 68, 459-475). Although this is less frequent, other starting materials can also be used, such as supernatants (S/N) of this Cohn fractionation, such as the S/N of Fraction IV or S/N of Fraction II+III, including an additional stage of purification, such as ion exchange chromatography, for example.
The purification methods that start with these Cohn fractionation intermediaries, with an albumin purity level of >90-95%, are normally based on the separation of polymers, aggregate material and other compounds such as lipoproteins, insolubilised by the effects of the ethanol present in the solution. For an optimum separation of these compounds it is advisable to add albumin of filtration aids (diatomaceous earth) and/or insoluble inorganic silicates (bentonites) or colloidal silica to the suspension. Optionally, according to the type of protein contaminants of the starting material, it can be advisable to add ion exchangers, maintaining conditions which prevent the adsorption of albumin.
Subsequently, elimination of ethanol from the solution is achieved by diafiltration in ultrafiltration equipment, which also allows the elimination of metallic ions and organic carboxylic acid anions such as citrate (Spanish patent P 2.107.390). This ultrafiltration also allows the albumin solution to be brought to the required concentration.
In addition, there is also the possibility of carrying out heat treatment of the solution (preferably heat shock between 56° C. and 63° C.), in order to denaturalise and insolubilise any thermolabile compound remaining in the albumin solution which could destabilise the final product. This heat treatment is also useful in reducing the activity of Prekallikrein Activator (PKA) which is one of the usual contaminants of albumin.
Different methods of albumin purification have been described, which represent variations to that described above, as regards sequencing of stages or adding additional purification stages.
Other methods cover albumin purification by ion exchange chromatography, beginning with fraction V or Supernatant of FrIV, such as U.S. Pat. No. 5,346,992 or EP 0367220. U.S. Pat. No. 4,288,154 suggests a process starting from Supernatant of Fr II+III, also by ion exchange chromatography, with a prior filtration stage in gel.
Human albumin solutions for therapeutic use obtained from plasma by the methods currently known appear in the form of a concentrated solution, generally with 15% to 25% of total protein, are hyperosmotic and cause a movement of fluids from the extravascular to intravascular compartment, or in the form of an isotonic solution of 3.5% to 5% total protein, which is isoosmotic with the plasma. The formulation of these solutions contains albumin, with a purity level of above 95% of total protein, and stabilisers.
When obtaining albumin solutions from human plasma pools, as with other biological products, the risk cannot be excluded of transmitting infection from viruses or other pathogenic agents. Despite this, current therapeutic albumin is considered a safe blood product in the face of this risk. This is especially due to the fact that albumin solutions are pasteurised (heat treatment of at least 10 hours at 60-63° C., generally in the final vial).
From 1941, the date that the first albumin solution from fractionating human plasma was prepared (Cohn et al. Chem. Rev., 1941, 28, 395), studies were begun to stabilise albumin solutions and in 1945 the stabilisation of albumin solution with N-acetyltryptophan, with sodium caprylate or with a mixture of the two was achieved (Scatchard G, et al, J. Clin. Invest. 24:671-676, 1945). The stabilisation of albumin solutions continued to be a subject of study (Finlayson J. S. et al, Vox Sang. 47:7-18, 1984); (Finlayson J. S. et al, Vox Sang. 47:28-40, 1984), but there have been no significant variations until today, this form of stabilisation being accepted by the regulatory authorities of the various countries (for example: European Pharmacopoeia; Human Albumin Solutions monograph 0255). Therefore, at present, commercial preparations of albumin in general contain as stabilisers, caprylate and/or N-acetyltryptophan, in order to avoid polymerisation during the process of pasteurisation and to ensure stability during the storage of the product until its expiry date.
In cases where the therapeutic effect of albumin occurs when capturing toxic compounds as a mechanism prior to its elimination, as in the treatment of hyperbilirrubinaemia, the efficiency of the treatment depends on the albumin having free binding sites.
In studies with models of hyperbilirrubinaemia in neonates, Ebbesen F. (Acta Pediatr. Scand. 71:85-90, 1982) establishes that N-acetyltryptophan powerfully displaces bilirubin, so reducing the efficiency of albumin stabilised with N-acetyltryptophan and Caprylate in contrast with albumin stabilised with Caprylate alone. Despite either of the two preparations being effective in protection against encephalopathy caused by bilirubin, the data seem to indicate that stabilisation with N-acetyltryptophan would not be suitable for treatment of hyperbilirrubinaemia, demonstrating the advantage of the stabilisation of albumin with Caprylate.
In other studies, also on the subject of the affinity and capacity of albumin to bind to various compounds, including Tryptophan and Caprylate, it has been established that there is competition between the various compounds for the albumin binding sites (Kragh-Hansen U. Biochem. J. (1991) 273, 641-644) and it is accepted that whilst Tryptophan binds to the albumin molecule at a specific site, Caprylate (octanoate) binds to the albumin molecule at various sites, causing a reduction in the capacity of albumin to bind to other compounds.
What is at present beyond all doubt is that these stabilisers bind, at least, to the albumin binding site II and so hinder the transporting function of pasteurised albumin, and can inhibit, at least partly, binding of the endogenous ligands (toxins, etc.) after the intravenous infusion of this albumin.
Attempts have been made in this direction to develop albumin production methods which avoid the use of N-acetyltryptophan and/or sodium Caprylate both during the purification process and in the final composition.
Patent WO2004071524 relates to an albumin preparation process which avoids the inclusion of stabilisers of the N-acetyltryptophan and Caprylate groups. For this we opt for a virus inactivation stage by treatment with Solvent-Detergent (SD) to replace pasteurisation. Treatment with SD has demonstrated great efficiency in virus inactivation with a lipid envelope, but it has no significant effect on the elimination of viruses without an envelope. Another disadvantage of this method is the need to eliminate SD inactivation reagents, which involves the inclusion of extraction stages which complicate, extend and make the process more expensive. Finally, as we have seen, the presence of PKA in albumin solutions is partly reduced by heating. As this patent does not cover heat treatment, it must also include a specific stage for the elimination of this, complicating this process even more.
U.S. Pat. No. 5,919,907 also concerns an attempt to obtain albumin with a greater binding capacity, by developing a process which avoids pasteurisation and thereby adding stabilisers of the N-acetyltryptophan and Caprylate groups. This process takes advantage of the biocide capacity of iodine compounds. Due to the affinity of albumin molecules for iodine and thus the capacity of albumin to neutralise its biocide effect, this method requires, according to said patent, the addition of an iodine compound in a sufficient amount to saturate the binding capacity of albumin and also allow sufficient free iodine to remain in the solution with biocide effect. In this respect, the reproducibility of the process does not seem easy to achieve, this aspect being fundamental to a stage of elimination of pathogens. Moreover, this method requires the elimination of the inactivation reagent, which also involves the inclusion of extraction stages which complicate, extend and make the process more expensive. Another aspect to be considered, not mentioned in said patent, is the presence of PKA in the albumin solutions. As in the previous case, when heat treatment is not performed, a specific stage must be included for the elimination of PKA, which complicates the process.